Human AF was donated from consenting adults during routine, planned cesarean sections under an IRB approved protocol (IRB approval agency: IRCM) in the operating room by aspiration. Donor qualification was performed under FDACFR 1271 guidelines. Donor qualification was certified following the review of the mother’s medical history, social history, physical examination, and raw product recovery information. Relevant communicable disease testing was completed, and the mother was reported to have negative/non-reactive results for CMV IgM Ab, Hepatitis B core total Ab, Hepatitis B surface Ag, Hepatitis C virus Ab, HIV-1/HIV-2 Plus O, HTLV I/II Ab,Syphilis screening—non-treponemal, Ultrio Elite HBV, UltrioElite HCV, Ultrio Eliter HIV-1/2, and WNV. All patients provided preoperative written informed consent. AF collections from 7 donors were used for this study. The average age of donors was 27.71 ± 1.9 years and the average gestational age was 38.3 ± .74 weeks. The average volume of unprocessed AF collected from each donor was 492 ± 140ml.

Full-term human amniotic fluid was collected in the operating room by aspiration during planned cesarean sections and shipped to the manufacturing faculty on ice. Upon processing, raw AF was centrifuged at 500g for 10 minutes. The supernatant was collected and centrifuged at 2000g for 20 minutes. The supernatant was then filtered through a.22µm filter to remove all large cell debris and large proteins. The processed product resulted in acAF comprising acellular extracellular components such as EVs and soluble proteins.

A starting volume of 16ml acAF was used for AF-EV preparations. To create the AF-EV preparation, acAF was further centrifuged at 100,000g for 3 hours at 4°C using an Optima XPN-90 Ultracentrifuge (Beckman Coulter) with a type 50.4 Ti fixed-angle titanium rotor (Beckman Coulter). The supernatant was removed, and the pellet was washed by resuspending in sterile DPBS and centrifuging a second time at 100,000g for 2 hours at 4°C. The final EV pellet was resuspended in 500µl DPBS and stored at -80°C until use.

EV-depleted supernatant was prepared by collecting acAF supernatant from the first centrifugation of AF-EV preparations, and centrifuging it at 100,000g for 16 hours. The EV-depleted supernatant was then collected.

Human peripheral blood mononuclear cells (PBMCs) were isolated from the blood of healthy consenting adult donors using a Ficoll density gradient (Ficoll Paque Plus, Cytiva). 6 donors participated in this study, and 60ml of blood was collected per donor. Blood was centrifuged at 1000g for 15 minutes in 50ml LeucoSep tubes. 1-1.2 million cells were seeded per well in a 24 well plate in RPMI media with 10% FBS and 1% Penicillin/Streptomycin per well in a total volume of 1ml. For PHA-stimulation 5µg/ml of Phytohemagglutinin-M (PHA) (Sigma 11082132001) was added to PBMCs in culture to stimulate T cell proliferation and activation. PBMCs were suspended in 1ml total volume which consisted of 800µl of media and 200µl of a treatment solution or a vehicle control. Treatment solutions added to PHA-stimulated cells included acAF, AF-EV preparations, and EV-depleted supernatant.

For T cell proliferation experiments PBMCs were stained with CellTrace™ CFSE dye (ThermoFisher Scientific C34554) prior to culturing them. CellTrace™ CFSE dye was diluted 1:1000 in PBS and warmed to 37°C. Diluted CFSE dye was added to PBMCs and incubated at 37°C for 20 minutes. RPMI media with 10% FBS was added to PBMCs and incubated at 37°C for 3-5 minutes to quench the remaining dye.

For the inital T cell proliferation experiments with acAF, to obtain a 10% concentration of acAF, 100µl of serum-free media and 100µl acAF were added to 800µl of media. To obtain a 20% concentration of acAF, 200µl of acAF was added to 800µl of media. 10% and 20% concentrations of acAF were added to experimental treatment groups in addition to the PHA to determine effective dose. An equal volume of serum-free media was added to the PHA only group and the non-stimulated control group. The media was changed at 4 days and was replaced with the same conditions. T cell proliferation was measured 4 days and 8 days after the addition of PHA, treatments, and control.

For the inital experiments with high concentrations of AF-EVs and acAF, the acAF was added at a 20% concentration (200µl in 800µl of media) and contained 4x10¹⁰ particles per 1x10⁶ cells, while the AF-EV solution was added at a 20% concentration (200µl in 800µl of media) and contained 1x10¹¹ particles per 1x10⁶ cells. An equal volume of serum-free media was used for PHA only group. T cell proliferation was measured 5 days after the addition of PHA, treatments, and control. Based on experience with initial experiments, an additional day was added to the early proliferation time point (day 4) to further amplify the percentage of proliferating cells responding to PHA and for the observation of immunosuppressive effects on a larger proliferating cell population at day 5. For T-Cell proliferation experiments, each PBMC donor was performed in triplicate wells and then averaged to create a singular data point.

For initial T cell activation experiments with acAF, (200µl in 800µl of media) were added to experimental treatment groups in addition to the PHA. An equal volume of serum-free media was used for the PHA only group and the non-stimulated control group. T cell activation markers CD25 and CD69 were analyzed 1 day and 3 days after the addition of PHA, treatments, and control.

For inital experiments with high concentrations of AF-EVs and acAF, was added at a 20% concentration (200µl in 800µl of media) and contained 4x10¹⁰ particles per 1x10⁶ cells, while the AF-EV solution was added at a 20% concentration (200µl in 800µl of media) and contained 1x10¹¹ particles per 1x10⁶ cells. An equal volume of serum-free media was used for the PHA only group and non-stimulated control group. T cell activation was analyzed 3 days after the addition of PHA, treatments, and control. For T-Cell proliferation experiments, each PBMC donor was performed in triplicate wells and then averaged to create a singular data point.

For dose response experiments, acAF and AF-EV preparations were diluted in PBS to obtain various nanoparticle doses: 4x10⁹, 2x10¹⁰, and 4x10¹⁰. First, the AF-EV preparations were diluted in PBS to obtain an equal concentration of nanoparticles/ml as the undiluted acAF (approximately 2x10¹¹ nanoparticles/ml)to create the first stock solutions for the 4x10¹⁰ dose. To create the other dose stock solutions, the first stock solutions of acAF and AF-EVs were each diluted 1:2 in PBS to obtain the second stock solution for the 2x10¹⁰ dose and 1:10 in PBS to obtain the third stock solution for the 4x10⁹ dose. Next, the volume of undiluted acAF which contained 4x10¹⁰ particles (175-200µl) was added to the media. Next, an equal volume of the AF-EV preparation first stock solution was added to the media to match the same nanoparticle dose. Next, an equal volume of second stock solution was added to the media to deliver 2x10¹⁰ particles, and an equal volume of third stock solution was added to the media to deliver 4x10⁹ particles. Lastly, an equal volume of supernatant was added to the media. All groups were treated with 5µg/ml PHA and an equal volume of PBS was added to the PBS group as a vehicle control. The proliferation studies were completed as described above and proliferation was measured at day 5. T-Cell activation studies were completed as described above and activation markers were measured on day 3.

Cells were collected at different time points, washed with PBS, and centrifuged at 350g for 5 minutes. Cells were resuspended in FACs buffer (Thermo Fisher Scientific) and incubated in antibodies for 25 minutes in the dark at room temperature. Following incubation, cells were washed twice with FACs buffer and analyzed with a Cytoflex (Beckman Coulter). 10000 events were recorded per sample. To analyze T cell proliferation and activation within the PBMC cultures, first CD3+ cells were gated. For T cell activation CD25+ and CD69+ cells were gated within the CD3+ population. For T cell proliferation experiments a CD3 APC (BD Biosciences 555342) antibody was used. For T cell activation experiments CD3 APC (BD Biosciences 555342), CD69 PE (BD Biosciences 654975), and CD25 FITC (BD Biosciences 555431) antibodies were used. For T cell viability experiments CD3 APC (BD Biosciences 555342) and Propidium Iodide PE (Thermo Scientific BMS500PI) antibodies were used. For EV uptake experiments CD3 APC (BD Biosciences 555342), CD4 APC (Biolegend 317415), CD8 APC (Biolegend 344721) antibodies were used. Isotype antibodies used were: Anti IgG1κ PE (BD Biosciences 555749), Anti IgG1 FITC (Thermo Fisher GM4992), Anti IgG1κ APC (BD Biosciences 555751), Anti IgG2aκ APC (BD Biosciences 555576).

Protein concentrations were calculated using a BCA Assay (abcam). 50µl of each sample was used and samples were run in duplicate. Samples were read on a microplate reader (Molecular Devices) at an absorbance wavelength of 562nm.

Nanoparticle tracking analysis of both acAF and the EV preparation were performed using a NanoSight NS300 instrument (NTA 3.4 Build 3.4.003). 10 µL of sample was diluted in 10 mL of Cell Culture Grade Water. The capture settings were modified to capture 5 video files with a capture duration of 30 seconds, using a camera level of 15, and a continuous syringe pump flow rate of 50. After completion of the script, the video files were analyzed with a detection threshold of 3. NTA post acquisition settings were kept constant between samples. The size and concentration profiles of each mode were then imported into Prism (Graph Pad, version 9.2.0) and superimposed.

Fluorescent NTA (fNTA) was performed using a ZetaView QUATT with ZetaView software version 8.05.12 SP1 (Particle Metrix GmbH,Inning am Ammersee, Germany). AF-EVs were labeled with anti CD63-Alexa 488 (NBP2-42225AF488) and anti CD81-DyLight 550 (NB100-65805R) (Novus Biologicals, Littleton, CO, USA) by adding 1mL of each fluorescent antibody to 20mL of sample containing isolated EVs. The fluorescently labeled EV samples were then incubated for 2h in the dark on ice. The samples were diluted by mixing deionized water filtered through a 0.2-µm syringe filter with corresponding volumes of sample. The fNTA was performed in scatter mode, 488/500 fluorescent mode and 520/550fluorescent mode. For scatter mode analysis, the ZetaView settings were adjusted to have a sensitivity of 75, shutter speed of100, cycles/positions of 2/11, frame rate of 30, maximum size of 1000, minimum size of 20, track length of 15 and minimum brightness of 20. Fluorescent mode analysis had similar parameters with the exception of an increased sensitivity of 8085. The size and concentration profiles of each mode were then imported into Prism (Graph Pad, version 9.2.0) and superimposed. This experiment was repeated using three independent AF-EV preparations.

AF-EVs were analyzed using a JEOL JEM-1230 electron microscope (JEOL USA, Inc, Peabody,MA, USA) in conjunction with formvar carbon-coated transmission electron microscopy grids (FCF400-Cu; Electron Microscopy Sciences,Hatfield, PA, USA). The copper carbon formvar grids were glow-dis-charged prior to loading an undiluted sample of EVs. The grids were then floated on 10mL of EV sample drop for 15 min, washed two times with water by floating on the drop of water for 30 s and then negatively stained with 2% uranyl acetate by floating on the drop of stain for 30 seconds. The grids were blot-dried with Whatman paper (Cytiva, Marlborough, MA, USA) and then imaged.

Unprocessed (raw) AF lysate was prepared by centrifuging unprocessed amniotic fluid at 1000g for 10 min to pellet cell debris and proteins. The pellet was lysed in 0.2mL RIPA buffer and centrifuged at 14,000g for 10 minutes. Protein was measured with a BCA assay for both raw AF lysate and AF-EV samples and the same amount of protein (80 ug/mL) was loaded for each sample. Capillary western was performed using 12-230 kDa prefilled plates (Bio-techne) with immunoassay and total protein detection. For total protein analysis a total protein detection module for chemiluminescence based assays (Bio-techne) was used. Separation time was 25 minutes, separation voltage was 375 V, primary antibody incubation time was 40 minutes, and secondary antibody incubation time was 30 minutes. Antibody signal was detected with HDR Chemiluminescence. Antibodies used were CD9 (Cell signaling technologies), TSG101 (Novus Biologicals), Annexin V (R&D Systems), and GM130 (Novus Biologicals).

200µl of media from PBMCs in culture was collected from each well and diluted 1:2 in serum-free RPMI media. Eve Technologies performed a multiplex immunoassay (Luminex, HDF-15) to analyze pro-inflammatory cytokine levels.

EVs were stained according to protocol described in Morales-Kastresana et al., 2017 with a few modifications (24). CellTrace™ CFSE dye (ThermoFisher Scientific C34554) was diluted 1:100 in PBS and warmed to 37°C. Diluted CFSE dye was added to AF-EV preparation at a 1:1 volume ratio and incubated at 37°C for 2 hours. Diluted CFSE dye was added to PBS at a 1:1 volume ratio as a control and prepared identical to the AF-EV preparation. 0.1% FBS in PBS was added to EV preparation and incubated at 37°C for 3-5 minutes to quench the remaining dye. The EV preparation was filtered through a Amicon Ultra-4 Centrifugal Filter Unit with Ultracel-100 membrane-100 kDa MWCO by centrifuging at 4000g for 15 minutes. The remaining EV solute was washed once with PBS and centrifuged at 4000g for 15 minutes. For cell culture experiments 1x10¹¹ particles from the EV solute were added to each well containing 1x10⁶ PBMCs stimulated with PHA and incubated for 3 days.

Data were expressed as mean± SD. Data points within each figure represent experiments with unique PBMC donors (n=3). The number of acAF, AF-EV, and AF-supernatant donors used within each experiment is indicated in the figure legends. Statistical analysis was performed by Prism software (GraphPad version 9.3.1) using one-way ANOVA with Tukey’s Test for multiple comparisons. Student’s t-test was used to analyze T cell viability. P<0.05 was considered statistically significant. The cell response to PHA stimulation may vary depending on the PBMC donor, therefore the T cell response was graphed as a percentage of the PHA vehicle control group average within the same experiment. Each experiment in this study was repeated with PBMCs from 3 separate donors.

To study the immunomodulatory effects of acAF in vitro, PBMCs were stimulated with PHA to induce cell proliferation for up to 8 days. 4 days following PHA stimulation, acAF suppressed T cell proliferation with both 10% and 20% concentrations ( Figure 1B ). Treatment with 10% acAF reduced the proliferating T cell population to 72 ± 9% of the PHA vehicle control group (p<0.01), while treatment with 20% acAF reduced the proliferating T cell population to 42 ± 9% of the PHA vehicle control group (p<0.001). There was no significant change in T cell proliferation in the group treated with 10% acAF treatment at the 8 day time point ( Figure 1C ). However, treatment with 20% acAF reduced the T cell proliferating population to 52 ± 17% of the PHA vehicle control group (P<0.05), suggesting a larger dose of acAF is required to reduce T cell proliferation over a longer time course. T cell viability was not affected by 20% acAF treatment at either time point ( Figure 1D ). A 20% concentration of acAF treatment was used for all further experiments since this concentration was found to be most effective.

To further explore immunomodulatory effects of acAF on T cells, the same cell culture model was used to study T cell activation. Upon activation, T cells overexpress membrane bound receptors CD69 and CD25. CD69 is an early-stage activation marker while CD25 is a later stage activation marker and has high expression levels about 2 days following activation until about 4-5 days (25). After 1 day of stimulation with PHA, there was a significant increase in both CD25+ and CD69+ T cells compared to non-stimulated control cells (p<0.0001) ( Figure 2A ). acAF reduced the population of activated T cells expressing both CD25 and CD69 (p<0.0001) but did not significantly change the populations of T cells expressing either CD25 or CD69 at this time point. After 3 days of stimulation with PHA, acAF treatment significantly reduced the CD25+ T cell population (p<0.01), the CD69+ T cell population (p<0.05), and the CD25+CD69+ T cell population (p<0.01) in PHA stimulated cells at the 3-day time point ( Figure 2B ). Overall, acAF treatment significantly reduced T cell activation and had a greater effect at later stages of T cell activation.

To investigate whether EV mechanisms are involved in acAF’s immunomodulatory effects on T cells, small EVs were further isolated from acAF using ultracentrifugation to create a concentrated EV preparation. Amniotic fluid contains high quantities of EVs and our group has previously published extensive characterization work for AF-EVs prepared with this isolation method (6) Protein concentrations were measured and particle analysis was performed for both acAF and AF-EV preparation samples. The small EV yield within the EV preparation was 1.61 x10¹⁰ particles/ml of unprocessed AF. The EV preparation had relatively low protein contamination (0.14 ± 0.03mg/ml), while the acAF sample had greater concentrations of protein (2.89 ± 0.26mg/ml). The EV preparation had over twice the concentration of particles as the initial acAF product (4.3 x10¹¹ particles/ml compared to 1.8x10¹¹ particles/ml in acAF); however, there was little variation in particle size between samples, with the mode particle sizes ranging from 135 to 172nM ( Figure 3A ).

AF-EVs were further characterized with TEM and fNTA. Transmission electron microscopy identified vesicles with a round morphology with an approximate size of 100nm ( Figure 3B ). fNTA analysis revealed the presence of exosome-associated transmembrane proteins CD81 and CD63 in the EV preparation. AF-EV nanoparticles were 64.9 ± 8.1% positive for CD81 and 70.1 ± 10.9% positive for CD63 ( Figures 3C, D ). Capillary western blot analysis confirmed the expression of EV marker proteins TSG101, Annexin V, and CD9 in the AF-EV preparation samples only, and relatively weak expression of Golgi body marker GM130 in both the AF-EV sample and raw AF lysate sample ( Figure 3E ). The raw AF lysate is not a pure cell product (it contains a mixture of vernix caseosa, extracellular matrix, proteins, and debris); therefore, cell organelle marker expression may be lower in this sample compared to pure cell lysate samples. Nonetheless, these results confirm our AF-EV preparation is enriched with EV positive markers compared to AF starting material.

To test the effects of AF-EVs on T cell proliferation and activation, a high concentration of AF-EVs (1x10¹¹ particles) from the EV preparation were added to PHA-stimulated PBMCs as an AF-EV treatment. In comparison, acAF treated wells had approximately 4x10¹⁰ particles added to each well. AcAF treated wells had approximately 0.3mg of protein added to each well while AF-EV treated wells had approximately 0.03mg of protein.

Concentrated AF-EVs had a more robust effect on both T cell proliferation and T cell activation than acAF. Both acAF and AF-EV treatment significantly reduced proliferation within the T cell population after 5 days of treatment (p<0.0001) ( Figure 4B ). However, AF-EV treatment suppressed T cell proliferation significantly more than acAF treatment (p<0.0001). acAF treatment reduced the proliferating T cells to 70 ± 3% of the PHA vehicle treated group, while the AF-EV treatment reduced the proliferating T cells to 36 ± 2% of the PHA vehicle treated group.

Additionally, concentrated AF-EVs reduced T cell activation significantly more than acAF treatment in CD25+ (p<0.0001), CD69+ (p<0.001), and CD25+CD69+ (p<0.001) cells ( Figure 4B ). The acAF treatment reduced the population of CD25+ T cells to 56 ± 5% of the PHA vehicle control group, and the AF-EV treatment reduced the population of CD25+ T cells to 28 ± 3% of the vehicle control group. The acAF treatment reduced the population of CD69+ T cells to 82 ± 4% of the PHA vehicle control group, and the AF-EV treatment reduced the population of CD69+ T cells to 64 ± 2% of the vehicle control group. The acAF treatment reduced the population of CD25+CD69+ T cells to 54 ± 1% of the PHA vehicle control group, and the AF-EV treatment reduced the population of CD25+CD69+ T cells to 30 ± 3% of the vehicle control group. There was a small difference between the CD25+ T cell populations in the AF-EV treated group and the non-stimulated control group (p<.05) compared to the difference between vehicle treated group and the non-stimulated control group (p<0.0001), suggesting concentrated AF-EV treatment maintained or restored CD25 expression to levels similar to those of naive T cells.

To further study the effects of AF-EVs on T cell function, a dose response was measured for both acAF and AF-EV treatment on PHA-stimulated PBMCs. Equivalent doses of nanoparticles from acAF and AF-EV preparations were added to compare the effects from each and to determine if other components of acAF other than EVs may be having an effect. EVs had a dose-dependent effect in both T cell proliferation and T cell activation ( Figure 5 ). There was no significant difference between the effects of acAF and AF-EV preparations with equivalent doses on T cell proliferation ( Figure 5A ). When the EV-depleted supernatant from acAF was added to PHA-stimulated PBMCs, this treatment also suppressed T cell proliferation, but this effect was less suppressive than that of the acAF ( Figure 5B ). There was no significant difference between the effects of acAF and AF-EV preparations with equivalent EV doses on T cell activation, and the acAF EV-depleted supernatant had no effect on T cell activation ( Figure 5 ). This suggests there may be a component in the acAF other than EVs which affects T cell proliferation, but not T cell activation.

To further study the effects of acAF and AF-EVs on immune cell activity, we analyzed pro-inflammatory cytokine release from PBMCs stimulated with PHA. PHA stimulation increased pro-inflammatory cytokine secretion from PBMCs, and both acAF and AF-EV treatment strikingly decreased PHA-induced release of pro-inflammatory cytokines GM-CSF, IFNγ, and IL-13 ( Figure 6 ). acAF and AF-EV treatment decreased extracellular GM-CSF levels by 80 ± 5% (p<0.01) and 98 ± 2% (p<0.001) respectively, IFNγ levels by 83 ± 11% (p<0.01) and 88 ± 5% (p<0.001), and IL-13 levels by 80 ± 7% (p<0.0001) and 92 ± 3% (p<0.0001). There was no significant difference between acAF and AF-EV treatment; however, average levels of all three pro-inflammatory cytokines were slightly lower with AF-EV treatment than with acAF. There was also no significant difference between control and PHA-stimulated cells treated with acAF or AF-EVs, indicating treatment restored extracellular pro-inflammatory cytokines close to non-stimulated baseline levels.

To investigate whether AF- EVs were being taken up by T cells, AF-EVs were labeled with CFSE and added to PHA-stimulated PBMCs. EV uptake was measured in T helper cells (CD4+) and cytotoxic T cells (CD8+) since these are the main subsets of T cells involved in activating the immune response. Both subsets of T effector cells, CD4+ and CD8+, exhibited AF-EV internalization ( Figure 7 ). AF-EVs were taken up in 48 ± 4% of CD4+ cells and 58 ± 2% of CD8+ cells after 3 days in culture. This suggests AF-EVs have a direct effect on T effector cells following internalization, rather than acting on T effector cells only indirectly via other cell types.